Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1275 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Classification | M5Product | 24 | Mar 25, 2026 | ||
| Continual Generalized Zero-Shot Learning | CUB | 24 | Feb 26, 2026 | ||
| Continual Generalized Zero-Shot Learning | AWA2 | 24 | Feb 26, 2026 | ||
| Deepfake Detection | DF 1.0 | 24 | Feb 26, 2026 | ||
| Object Detection | RainCityscape | 24 | Feb 26, 2026 | ||
| Semantic Segmentation | RainCityscape | 24 | Apr 28, 2026 | ||
| Personalized Image Aesthetic Assessment | PARA 1.0 (test) | 24 | Feb 26, 2026 | ||
| Scene Classification | NYU v2 | 24 | Mar 20, 2026 | ||
| Referring Image Segmentation | G-Ref UMD partition U (val) | 24 | Apr 2, 2026 | ||
| Multi-scale Image Generation | FFHQ 1024x1024 (test) | 24 | Feb 26, 2026 | ||
| Lossless Image Compression | CLIC mobile | 24 | Mar 31, 2026 | ||
| Few-Shot Learning | SUN | 24 | Feb 26, 2026 | ||
| Generalized few-shot learning | SUN | 24 | Feb 26, 2026 | ||
| Generalized few-shot learning | CUB | 24 | Feb 26, 2026 | ||
| 3D Reconstruction | ETH local feature benchmark Tower of London | 24 | Apr 8, 2026 | ||
| 3D Reconstruction | ETH local feature benchmark Gendarmenmarkt | 24 | Apr 8, 2026 | ||
| Classification | EuroSAT (test) | 24 | Apr 30, 2026 | ||
| 3D Human Pose Estimation | H36M (S9 and S11) | 24 | Feb 26, 2026 | ||
| Visual Grounding | ReferCOCO+ v1 (testA) | 24 | Feb 26, 2026 | ||
| Passage Ranking | TREC DL 2020 | 24 | Apr 7, 2026 | ||
| Adversarial Attack | VOC 2007 (test) | 24 | Feb 26, 2026 | ||
| Sentiment Analysis | IMDB Counterfactual (test) | 24 | Feb 26, 2026 | ||
| Oriented Object Detection | DOTA single-scale 1.0 (test) | 24 | Feb 26, 2026 | ||
| Gaussian Denoising | BSDS300 sigma=25 sRGB | 24 | Feb 26, 2026 | ||
| Aspect-based Summarization | SPACE (test) | 24 | Feb 26, 2026 |